01
Image
JPEG · PNG · WEBP
Two classifiers vote on whether the image was generated. If there are faces, we crop each one and run the same check. Camera physics and EXIF back things up when they can.
Is it real?
Drop in an image, video, audio clip, or text. TDFD checks it against a panel of detectors and shows you what each one saw.
Four kinds of input.
One kind of answer.
Upload an image, a video, a voice clip, or paste some text. You always get back the same thing: a score, a plain verdict, and the list of signals behind it.
02
Video
MP4 · WebM · MOV
We sample frames along the timeline and run each one through the image pipeline. A separate pass watches for the frame-to-frame wobble face generators tend to leave behind.
03
Audio
WAV · MP3 · FLAC · OGG
A wav2vec2 classifier handles the main call. Whisper features give a second read. Pitch, silence, and energy patterns flag what TTS and voice clones flatten.
04
Text
any UTF-8 paragraph
RoBERTa says whether it reads machine-written. GPT-2 measures how surprising the language is. Burstiness, rhythm, and LLM word tics round it out.
No single signal
decides it.
A verdict is the agreement between independent checks. One classifier reads the thing as a whole. Smaller forensic passes look for the specific fingerprints each medium tends to leave behind. If they line up, we say so. If they don’t, we say that too.
01
Classify
Pretrained models give the first read. One or two classifiers examine the artefact as a whole and output a probability. These carry the most weight because they've seen the most examples.
02
Forensics
Medium-specific artifact checks run in parallel. For faces, we crop and re-score each one. For audio, we probe pitch, silence floor, and energy envelope. For text, we measure burstiness and lexical rhythm.
03
Contextualize
Metadata and physics signals fill in what classifiers can't see. EXIF origin, sensor noise, chromatic aberration, and error-level patterns either corroborate or contradict the model scores.
04
Aggregate
Independent scores are agreement-weighted into a single probability. Where detectors agree, confidence rises. Where they diverge, the result shifts toward inconclusive rather than picking a side.
A detector with
an opinion.
i.
Show the working
Every answer comes with the signals that produced it, their individual scores, and a one-line note on what each one actually checked. You don’t have to trust the result. You can audit it.
ii.
Prefer learned over guessed
Trained classifiers do the heavy lifting. Hand-written heuristics fill the gaps where no good classifier exists, but they get labelled as heuristics and carry less weight in the final call.
iii.
Hedge when uncertain
Four possible verdicts: likely real, inconclusive, likely synthetic, highly likely synthetic. The middle two are the right answer more often than anyone wants to admit, and saying so beats pretending.
Get started
Ready when
you are.
Drop in a file or paste some text. You get a score, a plain verdict, and the full list of signals behind it — no account needed to try.